Iterative Feature Perturbation as a Gene Selector for Microarray Data

Abstract

Gene-expression microarray datasets often consist of a limited number of samples with a large number of gene-expression measurements, usually on the order of thousands. Therefore, dimensionality reduction is critical prior to any classification task. In this work, the iterative feature perturbation method (IFP), an embedded gene selector, is introduced and applied to four microarray cancer datasets: colon cancer, leukemia, Moffitt colon cancer, and lung cancer. We compare results obtained by IFP to those of support vector machine-recursive feature elimination (SVM-RFE) and the t-test as a feature filter using a linear support vector machine as the base classifier. Analysis of the intersection of gene sets selected by the three methods across the four datasets was done. Additional experiments included an initial pre-selection of the top 200 genes based on their p values. IFP and SVM-RFE were then applied on the reduced feature sets. These results showed up to 3.32% average performance improvement for IFP across the four datasets. A statistical analysis (using the Friedman/Holm test) for both scenarios showed the highest accuracies came from the t-test as a filter on experiments without gene pre-selection. IFP and SVM-RFE had greater classification accuracy after gene pre-selection. Analysis showed the t-test is a good gene selector for microarray data. IFP and SVM-RFE showed performance improvement on a reduced by t-test dataset. The IFP approach resulted in comparable or superior average class accuracy when compared to SVM-RFE on three of the four datasets. The same or similar accuracies can be obtained with different sets of genes.

Feature Selection for High-Dimensional Data

“…k-fold cross-validation is a common choice [42,44,48,57,58,61,68], as is holdout validation [43,46,52,55,59,63,69]. In fact, reviewing the recent literature one can find examples of the four methods described above.…”

Highly polygenic architecture of antidepressant treatment response: Comparative analysis of SSRI and NRI treatment in an animal model of depression

“…However this highlights the complexity and structure of the signal we are trying to detect. Robust Feature Selection (RFS) and Iterative Feature Perturbation (IFP) could also have been tested [Nie et al, ; Canul‐Reich et al, ]. A further possibility would be to use grouped LASSO which applies shared penalties to predetermined groups of features.…”

Feature Selection in DNA Microarray Classification

“…The authors in [58] introduced the iterative perturbation method (IFP), which is an embedded gene selector applied to four microarray datasets. This algorithm uses a backward elimination approach and a criterion to determine which features are the least important, relying on the classification performance impact that each feature has when perturbed by noise.…”

“…In fact, reviewing the recent literature one can find examples of the four methods described above. k-fold cross-validation is a common choice [42,44,48,57,58,61,68], as is holdout validation [43,46,52,55,59,63,69]. Bootstrap sampling was used less [50,60,66], probably due to its high computational cost, and there are also some representatives of leave-one-out cross-validation [62,64].…”